3'-specificity of Taq

[Dr. V. Blaschke]

Dear fellow researchers,

I think it is well established that Taq processivity in PCR depends on
tight 3'-binding of primers. Does anyone know exactly how many
mismatches at the 3'-end are necessary to completely abolish Taq
activity, i.e. would a 24-mer primer with a perfect overall match except
for a total mismatch of the three last 3'-bases work in PCR or not?

Any suggestions?

Thanks,

   Volker